Diagnosing children with signs of early puberty: knowing when to test and when to just monitor

KEYWORDS:

• Precocious puberty
• premature adrenarche
• genital hair of infancy
• premature thelarche

Specialists who see children referred because they have signs of precocious puberty have the challenge of trying to figure out which children have pathology and need testing and which can safely be monitored without extensive investigation. This topic was well covered by Drs Lughetti and Lucaccione in this publication in 2015 in a commentary titled ‘Precocious pubertal development: a challenge for pediatric endocrinologists’ [1], which included a discussion of the secular trends in the onset of puberty and its possible causes as well as the difficulties in deciding which children with true or central precocious puberty (CPP) should be offered treatment. My focus will be the large proportion of patients, mostly female, who do not meet the criteria of CPP (onset of progressive breast development before age 8 in girls) or ‘early puberty’ (often defined as onset between age 8 and 9), yet because of their pubertal signs generate concern on the part of their parents and primary care physicians. Which of these children need a diagnostic evaluation including hormone testing and bone age X-rays and which can be diagnosed as benign conditions based on just the history and physical exam?

One of the most common diagnoses made in children referred for early puberty is premature adrenarche (PA) [2,3]. The characteristic presentation is a child between the ages of 4 and 8 who develops pubic and/or axillary hair, often accompanied by axillary odor but not breast development in girls or penis and/or testicular enlargement in boys. One reason this condition seems to be so common, especially in the U.S.A., is that the mean age at which pubic hair appears has become earlier than in the past, particularly in females and African-American and Hispanic children [4]. Though not all children with PA are overweight or obese, excess weight is one of the key risk factors for early pubic hair development [2,5]. A study at my institution in Washington DC found that over a 2-year period, 62% of the 427 patients’ ages 5–9 referred for pubertal concerns were diagnosed with PA [6]. In contrast, in a study from a single center in Denmark of 449 consecutive puberty referrals from 1993–2009, only 49 (11%) were diagnosed with PA [7]. Thus, racial/ethnic background and environmental factors are likely important determinants of the frequency of PA.

How much testing do such children require? The one commonly abnormal test is an elevated dehydroepiandrosterone sulfate (DHEA-S) for age, which is expected since the etiology is an early increase in adrenal androgen secretion. Testosterone levels are invariably normal and 17-OH progesterone levels are typically normal but may be increased in the 2–4% who prove to have nonclassical congenital adrenal hyperplasia (NC-CAH) due to mild 21-hydroxylase deficiency [8]. Many believe that this justifies routine testing for NC-CAH, which sometimes presents with growth acceleration and other signs of androgen excess but which often is indistinguishable on clinical exam from PA. While there is no consensus that the milder cases benefit from treatment with hydrocortisone, which has the potential to slow growth and increase weight gain, making a diagnosis of NC-CAH would indicate the need for ongoing follow-up as a subset may benefit from treatment if signs of androgen excess progress.

Another important issue is the value of bone age X-rays which referring physicians often order even before the patient is referred, in the belief that it will identify patients who need more testing and/or treatment. In our study cited earlier, nearly half (121 of 266) of the patients with PA had a bone age X-ray done, and 30% had a bone age advanced by 2 years or more. This subset of PA patients were taller and more overweight than children with normal bone ages, yet only one patient had a diagnosis other than PA [6]. This calls into question the value of routine bone age X-rays in PA patients, which if advanced does not identify a subset of children with a high risk of a more serious condition, but often leads to physician anxiety and further unnecessary testing.

The child with fine genital or terminal hair in the labial or scrotal area appearing in the first year of life is another problem we see often. This was thought to be rare when first described in girls in 1992 [9] but has over the last two decades become increasingly common. At my hospital in Washington DC, we identified 71 patients (82% female) over a 4-year period with ‘genital hair of infancy’ [10]. In patients who had hormone testing, a mild elevation in DHEA-S was observed in about half, but increases in testosterone or 17-hydroxyprogesterone were not observed in females, though a few males had mildly increased testosterone consistent with the mini-puberty of infancy. Of those seen for follow-up, none had phallic enlargement, rapid progression in the amount of genital hair, or a growth spurt. Curiously, there were an additional 37 infants, all female, who had both fine genital hair and early breast development, also with a nonprogressive clinical course. It seems likely that genital hair of infancy is a very early onset form of PA with a similarly benign prognosis

Another familiar scenario is the girl who develops breast tissue in the first 2 years of life with no growth acceleration and who does not show progression over a period of observation. The diagnosis is almost always premature thelarche (PT), a common condition which is not associated with early menarche or short stature. In our study of 275 children referred before age 3 for signs of puberty, 156 or 57% were diagnosed with PT [10]. The mean interval between appearance of breast development and their endocrine consultation was 12 months, and breast stage was in nearly all cases Tanner stage 2, so typically there is a long period of little or no progression even before the child is seen. The question is whether testing is needed to rule out true precocious puberty and whether prolonged endocrine follow-up needed. A recent study from Italy found that of 450 girls seen for breast development starting before age 3, only 9 (2%) were found to have CPP during follow-up [11]. Furthermore, the results of stimulation testing with gonadotropin-releasing hormone analog (GnRHa) showed that luteinizing hormone (LH) often increased into the pubertal range in young girls who, during extensive follow-up, had a course consistent with PT and not CPP. While a very small number of very young girls with breast development prove to have CPP, most of whom have hypothalamic hamartomas, I suggest limiting hormonal testing to those girls who clearly show progression of breast development and/or rapid growth during a period of observation. This is an efficient strategy which will identify those girls who need testing and possible intervention while reducing unnecessary testing.

Some girls referred for early puberty have what appears to be breast tissue when they are examined in the sitting position, but when examined supine and with palpation, all that is found under the nipple is adipose tissue. In some cases, it can be difficult to distinguish breast from fat by palpation, but I have found that a repeat examination in 4–6 months will identify those girls with progressive CPP and that measuring gonadotropins and estradiol at the first visit is generally not necessary.

The final scenario I will discuss is the most puzzling: the prepubertal (or in some cases very early pubertal) girl who has one or more episodes of vaginal bleeding, often spotting lasting just a day. Again this causes great concern but in nearly all cases the physical exam is reassuring with no suggestion of infection, foreign body, trauma, or tumor. It is possible that in some cases, the source of bleeding is not the vagina but the bowel or urinary tract. We found by chart review of 24 such cases seen at our institution over a 5-year period that LH, follicle-stimulating hormone (FSH), estradiol, and pelvic ultrasound were normal in those tested [12]. What is characteristic of this condition, sometimes referred to as ‘premature menarche’ is spontaneous resolution after 1–4 episodes. While vaginal bleeding suggests some estrogen effect on the uterus, it is unexplained as to why there is no estrogen effect on breast development. If the physical exam is normal, reassuring parents that the problem is self-limited is best, though limited testing may be needed for girls with very anxious parents.

So, which patients referred for signs of early puberty need a full hormonal evaluation and possibly therapy with GnRHa? The key finding for girls is breast development starting before age 8 which progresses over a 4–6-month period of follow-up or which is already at Tanner stage 3, especially if accompanied by a rapid growth rate of 7–10 cm/year [13]. For boys, enlargement of both the penis and testes (>4 ml) also with rapid growth signals probable CPP. (Boys with only early testicular enlargement are less commonly referred since this is a subtle finding, but one which merits revaluation in 4–6 months and testing if there is progression.) The hormonal testing should focus on LH, FSH, and either estradiol or testosterone, and if results are equivocal, either close monitoring or GnRH stimulation testing is indicated. For the other scenarios mentioned earlier, parents can be reassured that the need for treatment is unlikely. Follow-up by the primary care provider should be considered, with reevaluation by an endocrinologist if there is growth acceleration or rapid progression of pubertal signs before age 8 (girls) or 9 (boys).

Declaration of interest

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.